Autocatalytic cycles, that is, ensembles of chemical compounds that can collectively selfamplify, are seen as putative key players in the origin of life, or more generally, in the physico-chemical emergence of Darwinian dynamics. Without directly displaying heritable variance in fitness (the necessary condition for evolution by natural selection) these objects nevertheless engage in rudimentary ecological dynamics that may pave the way toward evolution. Some cycles can’t run simultaneously, because they rely on incompatible chemical conditions, while others may interact in competitive or synergistic interactions. To investigate these dynamics, it is necessary to articulate physical, chemical and biological concepts, with the appropriate mathematical tools. The present internship proposal relies on such a framework that we have recently developed, introducing thermodynamic constraints in the detection and depiction of autocatalysis.